close

Вход

Забыли?

вход по аккаунту

?

Highly Reactive Intermediates from the Cocondensation Reactions of Iron Cobalt and Nickel Vapor with Arenes.

код для вставкиСкачать
Highly Reactive Intermediates from the Cocondensation Reactions
of Iron, Cobalt and Nickel Vapor with Arenes**
New Synthetic
Methods (80)
By Ulrich Zenneck"
In 1969, P.L. Timms reported the first preparative cocondensation reactions of metal vapors
with organic and inorganic substrates. The use of this technique in preparative chemistry soon
spread rapidly, but in recent years there has been less activity in this sector. If metal atom
reactions are not utilized primarily for the formation of new products, but for the synthesis of
highly reactive intermediates, a new synthetic strategy may be developed. Our aims are reaction sequences which, based on an effective cocondensation reaction, lead gradually and
selectively to new substance classes. This principle can be illustrated by the example of
the cocondensation products of arenes and iron, cobalt, or nickel vapor, which decompose
between -70 and -50°C. The classes of products accessible by this method extend from
clusters, through x-complexes, organophosphorus and organoboron cage compounds to pure
organic cycloaddition products.
1. Introduction
Many chemical reactions require an activation process. In
the case of a transition metal complex, this can mean that a
coordination site is opened up in a preceding reaction. The
substrate molecule can then directly attack at the center. On
the other hand, thermal or photochemical excitation of the
central particle gives rise to a highly energetic state, which is
particularly reactive. A special type of excitation is the production of free atoms by the vaporization of nonvolatile
elements. These exhibit enhanced reactivity since:
-
-
able to it. Either the desired reaction with the ligand results,
with formation of the monomeric complex ML,, or the metal atoms react with each other to give clusters M,L,, which
often then aggregate to a metallic state.['' Cluster formation
is generally the kinetically and thermodynamically favored
Thus, to obtain noticeable yields of the
monomeric complex, it is necessary to use large excesses of
+nL
They are free of ligdnds, which could kinetically hinder
any reaction ; substrate molecules encounter atoms with
each collison.
The atoms possess a high potential energy, so that reactions are mostly exothermic.
The combination of both of these factors suggests that the
reaction of free atoms with ligands should be carried out at
low temperature, and this is generally the case. In practice,
the region between - 100 and - 196 "C has proved to be the
best for preparative work.
The first work of this kind was carried out in 1963 by Skell
et al. on the vaporization products of graphite.''' In 1969,
Timms cocondensed transition metal vapors with organic
substrates for the first tirne.Iz1This initiated a rapid increase
in the amount of work being carried out in this field, which
appeared in several reviewsr3]and monographs.[4. 51
Neglecting the certain, gradual coupling of ligands to a
free metal atom, the simplified energetic course of a metal
atom reaction can be represented by Figure 1.
The vaporization of metals requires a relatively high vaporization enthalpy. In the case of transition metals, this
energy is, on the whole, higher than for the C-C bond.[61If
an atom is introduced into a cooled matrix or solution of
substrate molecules, there are two reaction pathways avail[*] Prw-Doz. Dr. U. Zenneck
Anorganisch-Chemisches Institut der Universitdt
Im Neuenheimer Feld 270, D-6900 Heidelberg (FRG)
[**I Reactive rr-Complexes of Electron-rich Transition Metals, Part 10.-Part
9: D. Hu, H. Schaufele, H. Pritzkow and U. Zenneck, Angew. Chem. fOf
(1989) 929; Angen. Chem. I n f . Ed. Engl. 28 (1989) 900.
126
Q VCH Verlugsgeselischujf mhH, 0-6940 Weinherm, 1990
Fig. 1. Simplified representation of the energetic course of a metal atom reaction (arbitrary units) AHy = vaporization enthalpy.
the ligand molecules (at least 10:1, better 50: 1). Therefore,
this method is limited to ligands which are readily available
and easily vaporized or highly soluble at low temperature.
The complex forming reactions are mostly strongly exothermic, due to the high potential energies of the metal atoms. A
clear tendency to give side reaction results, as can be seen in
the example of transition metal atom/alkyne reactions.181
This is not such a problem if, first of all, a highly reactive
intermediate complex is generated in a metal atom reaction,
and then allowed to react in situ to give the desired complex.
This simultaneously results in an increase in the usefulness of
metal atom reactions which can be worked with on a stoi-
0S70-0833j9Oj0202-Of26S 02.SOj0
Angew. Chem. Inf. Ed. Engl. 29 (1990) f26-137
chiometric basis at the second stage. The simplified energy
diagram of this reaction pathway is shown in Figure 2.
H
+nl
ML',Ly
Fig. 2. Energy diagram of a two step synthesis involving a metal atom reaction
and a highly reactive intermediate (arbitrary units) AH, = reaction enthalpy of
the second step.
The first examples of this type again came from the research groups of Skell and Timms, who allowed bis(arene)iron[*]to react at about - 60 "C with PF,. The result was
the simple substitution of an arene ligand with formation of
[(q6-arene)Fe(PF,),].[9s lo'
A systematic exploitation of the high energy available in
the intermediate step is only possible, however, if the pathway outlined in Figure 2 is broadened. The aim of the resultant reaction of the cocondensation product ML,
(ML, = reactive intermediate step A) should preferably not
be a thermodynamically stable product, but another reactive
particle which thermally decomposes below room temperature. Thus, from the partially extremely unstable cocondensation products A, one can obtain new, reactive, isolable
intermediates B which can be optimized to the requirements
of the desired synthetic aim. This step may be repeated, such
that further reactive but more easily handled complexes C
are synthesized from the isolable intermediates B. In the end,
a cascade A + B -+ C + D etc. of slightly exothermic reactions results from metal atom product A. Thereby emerged
the following hypotheses :
If the energy changes between the single steps are small,
then the reactions can be carried out selectively.
- The more intermediate steps, A, B, C, D etc. that can be
carried out, the greater is the variability of the resulting
chemistry.
In principle, this concept contains no limitations with respect to the metal or substrate which can be used.
The present review is intended to show some of the possibilities, for instance, of cocondensation products of iron,
cobalt and nickel vapor with arenes. We are particularly
interested in the cyclization reactions of unsaturated organic
molecules in the coordination sphere of reactive metal complexes. By using new intermediates, we are looking for novel
aspects of this reaction type, as compared with the known
systems.
-
2. Areneiron Complexes
The first reaction of iron atoms with an arene was reported
in the very first communication on metal atom chemistry by
Timms in 1969.['] The resulting bis(benzene)iron 1a exploded at about - 50 "C. As has already been described in Sec-
- a,
6"
cc
Fe(g)
+
2
Fe
2L
-6O'C
Fe
Rn
f \L
T Rnb
1 a-e
la: R,
Id: R,
=
H
= CF3
lb: R,
le: R,
=
Me
=
1.4-F2
lc: R,
=
1.4-Me2
tion 1, the thermally labile arene/iron cocondensation products can react with ligands at ca. -60 "C to give stable complexes of the type [(q6-arene)FeL,]. Ligands which have
been added by this route are: PF,,19. l o ] P(OR),, PR,, dil l and
olefins, cycloheptatriene, 1,3,5,7-cyclooctatetraenei1
2,T-bipyridyl (CC = metal -1igand cocondensation).['21
The structures of the cocondensation products of iron
atoms with arenes were disputed for a long time, for the
results of the investigations are dependent on the concentra-
Ulrich Zenneck was born in 1946 in Clausthal-Zellerfeld/ West Germany. After projessional
training as a dairyman and four years service in the federal border police, he studied chemistry
at the Universities of Clausthal und Marburg. In 1980 he received his Ph.D. from the University
of Marburg, under the supervision of Christoph Elschenbroich. He left Marburg for postdoctoral
studies with P. L. Timms in Bristol ( U K ) (1980/81) and moved to the University of Heidelberg
thereafter, where he finished his habilitation in 1988. He is now teaching there. His scientific
interests are centered on preparative organometallic chemistry, starting with metal vapors, and
redox reactions of transition metal complexes. His investigations of redox processes involve the
utilization of electrochemical and ESR and N M R spectroscopical characterization of pararnagnetic redox products.
Angew. Chem. Int. Ed. Engl. 29 (1990) 126-137
127
tions of the species involved, as well as on the reaction temperature. In principle, 1 : 1 and 2: 1 complexes, as well as
clusters can be postulated. In the case of the 2: 1 complexes q6:q6-, q6:q4- and q4:q4-structures have been discussed.[I3-' 81 The requirements of preparative metal atom
chemistry clearly favor the q6:q4-structure type for benzene
and toluene." 6 , 71 In contrast to this, the stable bis(hexamethy1benzene)iron has an q6:q6-structure, and can be prepared by classical methods.["]
In a number of cases, a three component reaction of Fe(g)
with arenes and ligands L has proved to be an alternative for
the formation of [(arene)FeL,] complexes." This is particularly true for benzene, since 1a usually gives only low yields
in substitution reactions with ligands.
2.1. Synthesis of Novel, Highly Reactive
Areneiron Complexes
Olefinic ligands can be rapidly substituted well below
room temperature in the case of cyclopentadienylbis(o1efin)cobalt and tricarbonylbis(o1efin)iron complexes,
which are isoelectronic with areneiron complexes.120.'I1
Thus, (q6-arene)bis(alkene)iron complexes seemed to be
promising candidates for novel reactive intermediates. On
the other hand, there was the opportunity to replace the
(q4-arene)-ligands of bis(arene)iron by naphthalene derivatives. This can offer additional bonding energy via better
rr-conjugation of the uncoordinated naphthalene ring. Accordingly an (q6-arene)(q4-naphtha1ene)iron complex of
type 3 should be more stable than an (q6-arene)(q4arene)iron complex of type 1. But it can still be assumed that
the q4-bonded naphthalene may be substituted by other
ligands as also (q6-naphthalene)-ligands can easily be
exchanged.[221Until now, structurally confirmed (q4-naphtha1ene)metal complexes have been confined to only two
examples with one 4d- and one 5d-metal respectively.[233 251 Naphthaleneiron complexes have thus far only
been obtained by matrix isolation in spectroscopic quantities." 71 Earlier reports of a [(naphthalene)Fe(CO),I complex[261were soon questioned.[271
Bis(to1uene)iron 1 b, freshly prepared by cocondensation,
is stirred in the temperature range - 80 to - 40 "C under
an atmosphere of ethene. A 56% yield of bis(ethene)tolueneiron 2 is obtained, calculated from the amount of
iron which is vaporized.[281Since it is impossible to direct all
of the metal vapor produced into the cocondensation
only about two-thirds of the metal atoms can react with the
toluene. The remainder form a metal layer on exposed surfaces of the apparatus. Thus, the chemical yield of the reaction is in excess of 80%. We usually vaporize 5- 10 g of iron,
or other 3 d-metals, over a period of 1-2 hours in our
"home-made'' apparatus, thus obtaining 10-20 g of 2 in less
than a day.[281Other complexes described herein are also
attainable in similar quantities. Cocondensation apparatus
of comparable efficiency are also commercially available. 2
decomposes in solution above -2O"C, and as crystals at
0°C. Long term stability can be obtained by storage at
- 78 "C under ethene. Decomposition products are the ligands and metallic iron. Despite its instability, 2 could be
characterized spectroscopically and by crystal structure
anaIysis.[281
Reaction of I-methylnaphthalene and 1b in the temperature range -80 to -4O"C, furnishes orange-brown (q4-lmethylnaphthalene)(q'-to1uene)iron
as a mixture of two isomers, 3 a and 3 b, in the ratio 5 :3.[291
-
2C2H4
Fe
////
lb
CllHlO
c
-Fe
Fe
+
2
30
3b
3a and 3b can be subjected to work-up by normal chromatographic techniques at - 30 "C. However, a successful
separation cannot be achieved by column chromatography.
Apart from their NMR spectra, there are no noticeable differences between 3a and 3b. They both decompose at ca.
0 "C, to the ligands toluene and I-methylnaphthalene, and
metallic iron.
The (1,4-dimethylnaphthalene)(p-xylene)iron derivatives
4a and 4b are analogous to 3, and also have very similar
properties. They are most conveniently synthesized in a three
component reaction between iron atoms and the ligands.
Again, the 5-8-q-isomer 4a is predominant over the 1-4-q-
La
Lb
249
128
complex 4b, but, in this case, the two isomers can be partially
separated by chromatography. (5-8-q-1,4-Dimethylnaphthalene)tris(trialkyIphosphite)iron 5 (orange) is obtained
from 1,4-dimethylnaphthalene,the relevant trialkylphosphites and iron vapor. Neither in 5a nor in 5 b have the
14-q-derivatives been observed. Unlike 3 and 4, both 5a
and 5 b are stable up to almost 150 "C, although they are far
more sensitive to oxidation.
50: R = Me
5b: R = Et
50.b
Reaction of 2 with 2-butyne and naphthalene affords
the brown complex (q6-hexamethylbenzene)(q4-naphtha1ene)iron 6, which is stable up to room temperature.[301
Angew. Chem. In!. Ed. Engl. 29 (1990) 126-137
GZ9Fe
+ 3 MeCCMe + CloH8
-
Fe
//
2
As with the other q4-naphthalene complexes,[24.2 5 s 291 the
bicyclic moiety in 6 is bent along the CI-C4 axis (Fig. 3). The
properties of this class of substances can best be understood
if one accepts that there is a very extensive decoupling of the
Fig. 3. Molecular structure of hexamethylbenzene(naphthalene)iron 6 in the
crystalline state.
complexed x4-system from the free #-system. The substances behave, both NMR-spectroscopically and in structural terms, about the same as would be expected for
benzoannelated 1,3-~yclohexadiene
complexes. Thus, the dihedral angle in the partly coordinated six-membered ring is
40 2" in the naphthalene complexes 5a and 6, as well as in
the cyclohexadiene complex 31 b (see Section 2.3). Also, the
'H- and 13C-NMR shifts for the corresponding nuclei of the
ligands of both substance types lie close to one another,['1'291 and the free rings of the bicyclic moieties are
influenced little by the metal.
Upon stirring a mixture of 1 b and 1,3,5,7-cyclooctatetraene (COT) variable amounts of the known complex cycl~octatetraene(q~-toluene)iron[~and the dinuclear complex p-cyclooctatetraenebis[(q6-toluene)iron]7 are formed,
depending on the concentration ratios used. 7 is obtained as
a mixture of different structural isomers.[311
Using an excess of 1 b gives 7 as the major product. Red
solutions of 7 stored at - 78 "C in aliphatic solvents, quantitatively decompose overnight to give cycloctatetraene(q6to1uene)iron and metallic iron. Solutions in toluene or benzene are stable up to 20°C, and then slowly form metallic
mirrors. Crystals of 7 are stable up to 70 "C, and then decompose in the same way as in solution.
2.2. Substitution Reactions with
Unsaturated Boron-Containing Heterocycles
As with 1, the initial product of the cocondensation of
arenes with iron, the novel reactive areneiron complexes 2,3,
4, 6 and 7 also give substitution reactions at temperatures
below room temperature with ligands such as phosphanes,
phosphites, diphosphanes, dienes and CO. These always
form complexes of the type [(arene)FeL,] in the first
step.[28-311If one disregards the more easily handleable
complexes which have been used here as educts, the reactions
do not, in principle, differ from the well-known analogous
reactions of derivatives of l.I9Since boron heterocycles possess a number of interesting
ligand properties,[321we decided to check whether the reactive complexes accessible from cocondensation reactions
would allow, should the situation arise, a route to areneiron
sandwich complexes with unsaturated boron-containing heterocycles as second ligand.
On allowing 3,4-diethyl-2,5-dihydro-2,5-dimethyl-l,2,5thiadiborole 8 to react with l a - l e or 2, the red compounds (q6-arene)(q 5-dihydrothiadiborole)iron9a-9e are
obtained as substitution
while 2,3-diethyl-1,4dimethyl-l,4-diboracyclohex-2-ene
and 1 b form the
orange-red sandwich complex 11 containing an (q6-diboracyclohexadiene) ligand by dehydrogenative c~mplexation.[~
51
-+
Fe
u Fe
7
Angen'. Chrm. Int. Ed. Engl 29 (1990) 126-137
9
Et
lb
+
Et
Me-p?-Me
u
10
-
--G+Fe
11
The reaction of the tetraethyl(methy1) derivative 12 a of
2,3-dihydro-IH-l,3-diborolewith 1 b follows a complex
p a t h ~ a y . 1 ~The
~ 1 desired substitution product, (q5-tetra-
ethyl(methyl)dihydrodiborole)(q6-to1uene)iron
13a (orange-red), can actually be obtained. What is particularly
interesting is the reaction of the thermally labile intermediate
15 with concentrated HCI, which is unusual in organometallic chemistry. Without this reaction, the only stable iron
complex isolated from 1 b and 12a is the hydrogenation
product 16. X-ray and neutron diffraction experiments carried out on single crystals of 13a[32,361
show that in this
complex, a pentacoordinated carbon atom (C-2) exists which
has enhanced reactivity-similarly to the electronically isovalent cyclopentadienyl cobalt complex of 12a.r371
Thus, 13a
can be deprotonated at C-2, quantitatively forming the corresponding complex anion.
The reactivity of the C2-H bond of complex 13a allows a
rapid and effective reaction with 1 b at low temperatures
which is much faster than the reaction of 1 b with 12a, but
129
leads to the same result. The combination of electrochemical c3'] and ESR-spectroscopical measurements on the reaction mixtures of 1b with 12a or 1b with 13a, furnishes results
from which a reaction hypothesis might be derived.
sible from 18 are complexes of the type [(benzene)FeL,] 19,
[(indenyl)FeL,H] 20 (or analogous cyclopentadienyliron
complexes), as well as CC-coupling products such as the
bridged ferrocene derivative 21.[431
A
FeCI,
HCI/Fe
\
From the measurements it emerges that the end product
13a is concomitantly the first intermediate of the reaction
sequence. A large number of oligodecker sandwich complexes containing the (p,q5-dihydrodiborolyl)(q6-toluene)iron
unit can be synthesized from 13a and its deprotonated prodU C ~ ,391
[ ~thus
~ , confirming the structures proposed for 14
and 15 on the basis of spectroscopic data.
Unlike 1b, 3 does not react with 12a, while reaction of 2
with 12a gives only small amounts of the substitution product 13a, along with the (q6-toluene)iron complex of a product formed by ethene insertion into 12a. An alternative route
to derivatives of 13 is the reaction of 2 with the diborafulvene
17f4']under an atmosphere of hydrogen. In this reaction, a
molecule of hydrogen is added across the exocyclic double
bond, whereupon the diborafulvene is transformed into a
2-isopropyldihydrodiborole derivative, and forms the sandwich complex 13 b.[421
Et
2
+
Et
I-\
Me-d
h
If
Me
Me
17
e + H,
-
Fe
13b
16
18 has proved to be a versatile reagent for these reactions,
and is rapidly accessible on a 20 g-scale, like 2 and 3, in a
three component reaction of iron vapor with the l i g a n d ~ . [ ~ ~ '
a
(CH,PR,),
19
18
20
21
Arenebis(trimethy1 phosphite)iron 22 is obtained almost
quantitatively from 1,[''] 2 or 3r291
by ligand exchange with
trimethyl phosphite. If it is allowed to react with additional
trimethyl phosphite and organic chloro compounds, it will
also lose its arene Iigand.[441
Gi-2
RCI
/
2 P(OMe),
2.3. Substitution of Arene Ligands
In benzenebis(trimethy1phosphane)iron 18," either the
arene, the phosphane, or the entire ligand sphere can be
substituted at room temperature, depending on the choice of
substrate. This is contrary to the reactions described in Section 2.2. Typical examples of the numerous products acces130
RCI = CHCI,.
R,
=
CDCI,,
MeCCI,,
H. Me, 1 . 4 - M e 2 .
PhCH2CI
23
1,3.5-Me3
Angru,. Chem. Int. Ed. Engl. 29 (1990) 126- 137
The formation of the dark green, paramagnetic iron complex 23 requires a Michaelis-Arbusov reaction, in which a
methyl group is removed, as well as an exchange of the arene
for additional phosphite ligands. According to the 31Phyperfine structure observed in the ESR spectrum (a, = 2.69,
a, = 3.06, a 3 = a4 = 3.63 mT; < g > = 2.063), the dimethylphosphonate ligand cannot be c3-bonded;[4s]rather, it
suggests an q'-coordination, which otherwise is only known
in thiophosphonate l i g a n d ~ . [ It
~ ~shows
]
good correlation
with the ESR data of 17-valence electron (VE) complexes of
the type [(q3-allyl)[P(OMe),],Fe].[471
2.4. Reactions with Alkynes
Fig. 4. Molecular structure of tolantris(tnmethy1 phosphite)iron 24a in the
crystal. Selected bond lengths [A]: Fe-C10 1.896(5),Fe-Cl1 1.850(6).
Alkynes belong to the most important group of substances
for cycloaddition reactions in the coordination sphere of
transition metals. The opportunity presented itself to investigate their reactivity with areneiron complexes.
A number of alkynes have been cyclotrimerized by the
complexes described in Section 2.1, in some cases at temper~
catalytic reatures lower than room t e m p e r a t ~ r e']. ~These
actions can be of use as an effective source of further, reactive areneiron complexes, which can be converted into new
alkyne complexes.1481
Reaction of 2 with alkynes at - 20 "C, followed by addition of trialkyl phosphite to the reaction mixture at less than
0 "C, leads to the the intensely blue-violet alkynetris(trialky1
phosphite)iron complexes %a-24f in fairly good yields.
R
RCCR
2
-
P(OR'),
-
-20°C
0°C
R
(R'O),P'
R'
2La: P h
2Lb: P h
Me
2LC:
COOMe
Me
2Ld:
2Le:
COOMe
Et
SiMe,
Me
Et
2Lf: SiMe,
wR
Fe,
1 '*'"P(OR')S
P
(OR'),
240-f
Et
Hitherto, 24a and 24b were only available in small
amounts by another route.[491Both the I3C-NMR signals
for this series of complexes (S(13C) for the sp carbons lies
between 178 and 206, cf. Ref. [SO]) as well as the results of
the crystal structure analysis of 24a (Fig. 4) indicate that the
alkynes in these compounds use all four n-electrons in their
bonding to the
Thus, they can be viewed as 18VE
complexes, which is reflected in their thermal stability.
Reaction of bis(trimethylsily1)acetylene with 2 affords the
two bis(p-alkyne)Fe, clusters tetrakis[bis(trimethylsilyl)acetylene]diiron(Fe = Fe) 25 (dark brown)15 and tris[bis(trimethylsilyl)acetylene]bis(trimethyl ph0sphite)diiron 26 (deep
violet).13'I While diamagnetic 25 can be synthesized just as
well from 3, paramagnetic 26 (two unpaired electrons) is
Angew Chem In! Ed. Engl. 25 (1590) 126-137
only accessible via the route using 2, since the trimethyl
phosphite must be added at between - 20 and 0 "C.
R
RCCR
2 - --2 0 ° C
-20°C
L
J
R=SiMe,
-10°C
i
Jl
R
I
P(OMe),
R
R
<
RCCR
[I
Fe&e
R
R
R
25
26
Since both dinuclear complexes are electronically isovalent (the terminal alkyne ligands of 25 and 26 function as n4
ligands as in 24), the structural (Fe-Fe distance: 25,
2.46 A;[''] 26, 3.08
and magnetic"31 differences
between them are noteworthy.
25 is the first, structurally characterized, homoleptic
alkyne metal cluster. The practically quantitative route from
2 or 3, directly illustrates the great advantage of a "roundabout" approach to the synthesis using reactive intermediates. The product formed by the direct cocondensation of
iron atoms with bis(trimethylsily1)acetylene has been de-
U
Fig. 5. Molecular structure of {(q6-toluene)(qs-1-(q6-toluene)-2,3,4,5-tetrakis(methoxycarbonyl)ferrole]}iron 28 bin the crystal. The toluene ligand of Fel
is rotationally disordered.
131
scribed as a non-stoichiometric substance,[541which also obviously contains some of 25.
The C-C linkage of two alkynes in the catalytic cyclotrimerization by areneiron complexes leads presumably
first of all to ferrole complexes of the type 27. The likely
existence of these can be demonstrated by allowing an excess
of 1 b to react with an alkyne. Dark green, complex-stabilized ferroles 28 are formed (Fig. 5). Tetraphenylcyclobutadiene(to1uene)iron 29 (Fig. 6), which is a o-7~rearrangement
3 (2.lb)
-
28a.b
29
30
28a: R=Me
28b: R=COOMe
R=Ph
R=Ph
2 RCCR
31a-c
R=COOMe
31 a : n=3
31 b: n=4
31 C : n=6
Fig. 6. Molecular structure of (q4-tetraphenylcyclobutadiene)(qb-toluene)iron
29 in the crystal. Selected bond lengths [A]: Fe-C(C,R,) 2.002-2.050; Fe-C(toluene) 2.036-2.138.
product of the ferrole, and the hydrogen addition product
tetraphenylbutadiene(to1uene)iron (30, Fig. 7) are formed
together from 1 b and t ~ l a n . [The
~ ~ ]origin of the hydrogen
for the formation of 30 has not been explained in this instance. Hydrogen addition reactions have also been observed with other derivatives of l.'551
clohexadienes are formed in the coordination sphere of the
metal, and bound as ligands, as in the analogous reactions of
cyclopentadienylcobalt complexes.[561However, contrary to
the cobalt system, the reaction of the iron complexes (3 is most
suitable in this case) is strictly stereoselective. For example,
upon addition of cyclic mono- or diolefins, the olefinic
ring inserted into complexes of the type arene(cyc1ohexadiene)iron 31 are always found in the exo-p~sition.[~~]
The
endo-isomers cannot be detected NMR spectroscopically. A
reason for the higher stereoselectivity of the iron complexes
could primarily be the low reaction temperatures usedmaximally room temperature-while for the cobalt complexes more than 70 "C is required.[561Decomplexation of
the ring bound to the metal in 31 and other related complexes
is simple. After addition of Fe3@-saltsto solutions of the
complexes, the ligands can be almost quantitatively isolated.
The cyclotrimerization products of alkynes have also been
fixed directly as complex ligands, and the reaction using
1,5-cyclooctadiene (COD) affords arene(cyc1ooctadiene)iron derivatives 32 in high
3
Fig. 7. Molecular structure of (q4-tetraphenylbutadiene)(q6-toluene)iron 30in
the crystal. Selected bond lengths [A]: Fe-C(C,R,H,) 2.060-2.113; Fe-C(toluene) 2.007-2.1 15.
29 and 30 form a pair of electronically isovalent complexes. The Fe-C distances differ only very little, while other
properties are markedly different. Thus, for example, 29 can
be reversibly oxidized and reduced, which is very unusual for
a mononuclear areneiron complex, and it only starts to thermally decompose at about 500 0C.[441
30, however, like most
other (arene)FeL, complexes, is air sensitive and decomposes at 210°C.
The important synthetic value of reactions of reactive
areneiron complexes with alkynes is shown by the [2 + 2 + 21cycloaddition of two alkynes with one olefin. Here, 1,3-cy132
+
3 RCCR'
R
320: H
32b: M e
32c: Et
32d: H
R'
H
Me
+
1.5-COD
-
R' Fe
#
32~1-d
Et
tBu
In this case, the entire ligand sphere of the iron has been
exchanged. Hence, for formation of benzene-, 1,3,5-trialkylbenzene- or hexaalkylbenzene-complexes, one can also start
with tolueneiron complexes 2 or 3, both of which are readily
accessible. Hexamethylbenzene(naphtha1ene)iron 6 (see
Section 2.1) offers the additional possibility of carrying out
such a synthesis stepwise. Firstly, the areneiron fragment is
formed from 2 and the alkyne between - 20 and + 20 "C,
Angrw. Chem. Inl. Ed. Engl. 29 (1990)126-137
and then the naphthalene is slowly replaced at room temperature by the desired ligands.
3
n +CgP
+
0/20"C
33
n
2.5. Reactions with tevt-Butyl Phosphaacetylene
terr-Butylphosphaacetylene 33[571was first used in metalcatalyzed cycloaddition reactions in 1986.[581Since then its
chemistry has developed exceedingly quickly.[s9.601 Since the
ligand sphere of the reactive areneiron complexes can be
completely exchanged for alkynes, we hoped that addition of
phosphaalkyne 33 would lead to formation of larger
oligomers at the iron center.
The reaction of 3 with 33 in molar ratios from 1 :2 to 1 : S
between 0 and 20°C yields the cyclodimer of 33, which is
bound as a ligand in 1,3-diphosphete(toluene)iron 34
(orange) and the sandwich complexes 35 (olive), 36 (green),
and 37 (green)[61* containing 1,3-diphospholyl- and 1,2,4triphospholyl ligands. The formation of these ligands must
also involve cleavage of the P-C triple bond in 33.
3a
*
=
39
LO
~ B u
38[66]and 39 are thermally stable 16VE complexes, which
are clearly distinguishable in this respect from other 16VE
complexes of iron.[671In 38 (Fig. 8), as in 34 and 35, there is
\
\
Fig. 8. Molecular structure of (PCCMe,),Fe 38 in the crystal
34 is a diphosphorus analogue of the cyclobutadiene complex 29; the properties of the two complexes manifest their
close relationship. 36 and 37 can be regarded as penta- and
hexaphosphaferrocene derivatives respectively. However
one should distinguish 36 from the similarly green pentaphosphaferrocene derivative containing a cyclo-P, hgand,
which has recently been described by Scherer et al. r631
Although one must accept that 36 is formed via an exceedingly complex reaction pathway, nevertheless five complete
molecules of 33 react with one molecule of 3 in the formation
of both five-membered rings. For 35 and 37, things are more
complicated. In these compounds, the ratio of the P/C units
of the phosphaalkyne is not 1 :1. In 35, there are four phosphorus atoms and five C-R fragments, while for 36 the ratio
is 6:4. This has to be interpreted in terms of an intermolecular exchange of these groups, at least in the early stages of the
isolated complexes.
By using a very large excess of phosphaalkyne in the reaction of 3 with 33 (which offers no problems, due to its
improved accessibility[641), the consumption of about 9
moles of 33 per mole of 3 is observed in less than a day at
room
Aside from 34 to 37, which are present in lower proportions than when using smaller amounts of 33, a product
mixture is now obtained which was shown by medium pressure chromatography to contain at least ten other iron-containing fractions. After repeated MPLC of this mixture, the
cornpiexes (PCCMe,),Fe 38 (dark brown, paramagnetic),
(PCCMe,),Fe 39 (olive brown, paramagnetic), and
P,(CCMe,),HFe 40 (red, diamagnetic) were isolated in
small quantities and characterized.[65,
Angm.. C'hcm. In!. Ed. EnKf. 29 (1Y90) f26-137
a diphospheteiron fragment which is bound to a bicyclic
tetramer of 33 in a complex, The interaction of the tetramer
with iron takes place via a 1,3-diphosphaallyl grouping and
a P-Fe o bond. In 39 (Fig. 9), the ligand is formed by a
/
Fig. 9. Molecular structure of (PCCMe,),Fe 39 in the crystal
polycyclic heptamer of 33. Again, there is a coordinating
P,(CCMe,), ring, but this is joined to a (PCCMe,), cage via
a o-bond between C6 and PS, and is bound as a 1,3-diphosphaallyl ligand. The cage is bound to the metal by a 2-phosphaallyl on the one side, and on the other side a coordination
site is occupied by the free electron pair of the phosphorus
atom PS. The skeletal framework of 40 can be described as
a ferrahexaphosphapentaprismane derivative (Fig.
in
which the P,(CCMe,),H moiety behaves as a tetradentate
133
i-
Fig. l l . Molecular structure of P,(CCMe,), 42 in the crystal
Fig. 10. Molecular structure of P,(CCMe,),HFe 40 in the crystal.
P, ligand. In contrast to 38 and 39, 40 has a closed shell,
3 8VE structure. The phosphirene ringc6*]P4C4C5, which
juts out of the molecule, is only bound via P4 to the framework and the metal. Thus, reaction of a larger excess of 33
with the iron complex 3 can lead to up to seven units of the
phosphaalkyne being coupled around the metal to give novel
types of cage structures. All attemps at the direct removal of
the organophosphorus skeletal frameworks from the complexes have thus far proved unsuccessful. However, heating
of the reaction mixture under vacuum at 100 "C, on completion of the above-mentioned reaction met with some success.
Thus, aside from the iron complexes 38 to 40, small amounts
of metal-free organophosphorus skeletal compounds can
also be obtained.
3
+
n+C-P
0/20"C
100°C
Fig. 12. Molecular structure of (PCCMe,),H,O 45 in the crystal.
L3
L2
0
L5
(PCCMe,), 41 is a polycyclic pentamer of 33,16,]the structure of which can be derived from that of the tetraphosp h a ~ u b a n e ' ~by~ replacing
]
one corner by a three-membered
ring, formed by insertion of a PCCMe, unit. A structurally
analogous hydrocarbon is known,[''] but not an organophosphorus compound or a polyphosphide. P,(CCMe,), 42
(Fig. 11) belongs to the same structural group.[661
P,(CCMe,),H 43[651is a heteroanalogue of homopentaprismane.[711Like 40,which also has a pentaprismane structure,
43 contains an additional hydrogen atom. P,(CCMe,),H 44
134
I
--
33
L.4
is formally derived from 43, which has been enlarged by an
additional unit of 33. The structural form of 44 is, to our
knowledge, new. Parts of the framework have similarities
with organopolyphosphanes of the type P,R, .f7'1
Tricyclic (PCCMe,),H,O 45 (Fig. 12)[661has far less internal bonding than the compounds 41 -44. Here, we are deal-
ing with an oxidation product, which has, in addition, incorporated four hydrogen atoms. The pentaphosphaferrocene
derivative 36, in particular, could be considered to be an
educt of 45. Not only is the number of PCCMe, units the
same in each, but also the relative orientation of the two
five-membered rings. Similar correlations can be drawn between 39 and 41, as well as 40 and 43. In both cases, only one
diphosphete-iron fragment has to be removed from the complexes, such that on formation of the new structures, the
remaining frameworks can be stabilized in their most saturated and closed conformations. An explanation to this
problem has been sought just as intensively as the principal
question, of how these systems organize themselves around
the metal center.
3. Arenecobalt Complexes
Arene/cobalt condensates have already proved useful in
the formation of bis(2,2'-bipyridyl)~obalt,'~~]
of reactive
organometallic slurries [741 and of catalytically active solut i o n ~ . [However,
~~]
up till now, it has not proved possible to
attach the arenes as x-ligands to the cobalt atoms, to form
cobalt complexes.
Angew. Chem. Inr. Ed. Engl. 29 (1990) 126-137
Reaction of the cocondensates 46a to 46e (arene =
benzene, toluene, p-xylene, trifluoromethylbenzene 1,4difluorobenzene) in the temperature range - 100 to -20 "C
with alkynes [2-butyne, bis(trimethylsilyl)acetylene] yields
both alkyne(arene)cobalt 47 (green, paramagnetic) and palkynebis(arenecoba1t) 48 (red, diamagnetic) in variable ratios, dependent on the type and concentration of the educts
used.[761
L6
.."
-a R
+ RCCR
CO
I E
.'n
L7
Me
Me
Me
Me
Me
SiMe,
hCO--'
proof of the alkynes using all four 7c-electrons in their bonding with the metal.
Chemical and electrochemical oxidation experiments on
47 a and 47f, to form 47 a@and 47 f@, have, up till now, been
unsuccessful. The lifetime of the predicted 18 VE cationic
complex 47@is so short, even at -6O"C, that only clean,
irreversible oxidations are observed.
Thermal decomposition of the mononuclear butyne complexes affords hexamethylbenzene, together with small
amounts of the corresponding dinuclear complexes. These
also yield the cyclotrimer of 2-butyne at higher temperatures.
Reaction of 46 b, not with alkynes, but with the N-C triple
bond of acetonitrile, leads to formation of small amounts of
the dark red trimetallatetrahedrane (pi,-benzy1idyne)tris(toluenecobalt) 49,[761and much metallic cobalt. 49 is not
observed in the thermal decomposition of 46 b.
68
H
Me
1 .4-Me2
CF,
1,4-F2
Me
The benzene complex 46a reacts with 2-butyne to give
47a. For the toluene/butyne system, the formation of 48 b is
strongly favored. Addition of excess butyne to 46 b increases
the amount of 47 b, so that the reaction path outlined in the
scheme is assumed. In the case of toluene/C,(SiMe,),, the
formation of 47f is favored, and 48f is formed in trace
amounts only. As in the toluene system, the dinuclear complex 48d is again favored in the trifluoromethylbenzene/butyne system, and the corresponding mononuclear complex
47d can only be identified spectroscopically. In the reactions
of butyne with p-xylene and with 1,4-difluorobenzene, the
respective derivatives of 47c and 47e, respectively, can be
observed by ESR spectroscopy. Separation of the reaction
products is achieved by fractional vacuum sublimation. It
must be accepted that certain losses arise during this process
(47 a-47 f slowly decompose above room temperature).
While the dinuclear complexes 48 are analogous to
[CO,(CO),(p-alkyne)]
there was, at first, no other example comparable to complex 47.[781
The crystal structure analysis of 47 a (Fig. 13), and interpretation of the ESR
data for the series of compounds,'761afforded unequivocable
49 is closely related to 48, and results from the formal
replacement of a C-R fragment by an isolobal arenecobalt
unit.
4. Arenenickel Complexes
Arene/nickel cocondensates also behave similarly to the
corresponding iron and cobalt compounds 1 and 46,respecti~ely.[~".
4, s] Upon thermal decomposition of (toluene),Ni
50[791at about -5O"C, very finely divided nickel with a
hydrocarbon coated surface is obtained, which can be used
for a number of reactions.[*']
When the decomposition of 50 is allowed to take place in
the presence of phosphites, [P(OR),],Ni is formed.1791In the
presence of ethene, (C,H,),Ni 51 a[''] is obtained, which is
also accessible via classical routes.IBZ1Owing to the small
amount of time required, and the use of very simple chemicals, together with the availability of a cocondensation apparatus, the route to derivatives of 51 via 50 is an interesting
alternative to the synthesis of 51a from cyclododecatrienenickel.
51a: R=H
51b:
Fig. 13. Molecular structure of (q6-benzene)(2-butyne)cobalt 47a. Selected
bond lengths [A]: CwC8 1.885(4); Co-C9 1.901(5).
Angew. Chem. Inr. Ed. Engf. 29 (1990) f26-137
R=C,H,
R
51a.b
The synthesis of 51 a from 50 is complete in less than one
working day, analogously to the synthesis of 2 from 1b. Even
135
the yields and amounts synthesized correlate very well (102 0 g per run). Reaction of 50 with hex-I-ene instead of
ethene furnishes tris(hexene)nickel51 b, which thermally decomposes above -30 "C. The characterization of 51b has
not yet been completed.
Reaction of 51 b with the dihydrodiborole derivative 12 a
(cf. Section 2.2) in a ratio of greater than 1 :1 in the presence
51b
+
12a
1
:
<I
The areneiron system, in particular, has proved its versatility in the results which have been presented and summed
up in this review. The initially formulated aim to convert a
few highly reactive cocondensation products gradually and
selectively via further reactive, but isolable intermediates inLO a number of novel compound classes, could, in some
cases, be achieved. For example, in the reaction sequence
Fe,,,,, + Fe,,, -+ 1 b 2 + 25, a total yield of 50% can be
obtained (calculated from the amount of iron vaporized);
20 g of 25 can be prepared in two days from elemental iron.
Pieces of nail or waste sheet steel have proved to be a far
better source of iron for metal vaporization than pure metal
powder, which tends to liberate too much gas when heated
under vacuum.
The intermediates are so reactive that in several cases the
entire coordination sphere of the metal complex is exchanged, even under mild conditions. Thus, novel types of
structures with an organoelement framework can be built
around a metal, using the appropriately reactive educts.
On searching through the literature of metal atom reactions conducted on a preparative scale and the corresponding matrix spectroscopy, it would seem that stable primary
products are in the minority. Thus, we feel confident that we
have revealed a much broader scope of application for highly
reactive cocondensation products in systematic syntheses,
than was hitherto ever thought possible.
-
-3O'C
of free hexene led to rapid polymerization of the hexene
above - 30 "C. Ethene polymerization can also be observed
in reactions of the ethene complex 51a with 12a or 12b in a
molar ratio greater than 1 : 1 below room temperature. However, the yields of polymers are clearly lower than with hexene.[831This may be caused by the olefin concentration, since
the reaction was conducted at normal pressure.
Reaction of 51 a and 12 b in a ratio of 1 :1.5 at room temperature for several hours leads to formation of the novel
green compound bis(diethyltrimethyl-2,3,5-tricarbahexaborany1)nickel 52,cS41 the parent compound of a series of
oligodecker sandwich complexes 53a to 53c, which are obtained in only small amounts via this reaction route.
1
Y
\
Ni
<4F
0
I
52
Y
53a: n = l
53b: n=2
5 3 ~ n:= 3
I
53 a-c
The same reaction can be optimized with regard to the
yield of the known tripledecker 53 a.1851
The educts, in a ratio
of 1:3, are heated to over 160"C, without solvent, at which
point, a little air is introduced to initiate partial
The higher oiigodeckers 53 b and 53c can be detected, although their amounts are still small. Qualitative MO considerations show that the qs-I ,3,5-tricarbahexaboranylligands
can be regarded as being isolobal with cyclopentadienyl ligands. 52 would then be a nickelocene analogue. In fact, all of
the properties which are related to the electronic structure of
the molecule can easily be understood in terms of this hypothe~is.~
This
~ ~ ]holds true, for instance, for the magnetism-which shows Curie-Weiss behavior of two unpaired
electrons for both compounds and for the electrochemistry.[881However the ESR data for the reversibly formed
19 VE-cation 52@are clearly different from that for the nickelocenium ion.1891The reason for this lies in the Jahn-Teller
activity of the nickelocenium, which does not appear in 52
for reasons of symmetry.
136
5. Conclusion
I would like to thank m y co-workers Dr. A . Funhof; D. Hu,
Dr. H. Schaufele, C. Tolxdorff, and Dr. L. Vasquezfor their
care, commitment and enthusiasm shown in carrying out this
work. The assignment of particular projects was as indicated
in the literature. Dr. H. Pritzkow performed all of the crystal
structure analyses which appear in this review. Reactions involving boron heterocycles were the result of strong collaboration with Professor Dr. W Siebert, Heidelberg, and his
co-workers Dr. G. Brodt, Dr. D. Biichner, Dr. L. Suber, Dr.
.I-K. Uhm, and Dr. J. Zwecker. I would like to thank them all
for their outstanding cooperation, and in particular Prof. Dr.
W Siebert for giving me all the necessary support for such a
project. Finally, I am most gratejul to the Stiftung Volkswagenwerk, the Land Baden- Wiirttemberg, and the Deutsche
Forschungsgemeinschaft ( S F B 247) for their generous financial support of this work.
Received: July 20, 1989;
supplemented: September 18, 1989 [A 748 IE]
German version: Angew. Chem. 102 (1990) 171
[I] P. S . Skell, L. D. Wescott, Jr., 1 Am. Chem. Soc. 85 (1963) 1023; P. S.
Skell, L. D. Wescott, Jr., J. P. Goldstein, R. R. Engel, ibid. 87 (1965) 2829.
[2] P. L. Timms, Chem. Commun. 1969, 1033.
[3] a)P. L. Timms, Adv. Inorg. Chem. Radiochem. 14, (1972) 121; b)P. S.
Skell, J. J. Havel, M . J. McGEnchey, Acr. Chem. Res. 6 (1973) 97; c) K. J.
Klabunde, Ace. Chem. Res. 8 (1975) 393; d) P. L. Timrns, M. Moskovits,
G. A. Ozin in M. Moskovits, G. A. Ozin (Eds.): Cryochemisrry, WileyInterscience, New York. 1976; e ) P. L. Timms, T. W. Turney, Adv.
Oganomet. Chem. 15 (1977) 53; f) G. A. Ozin, S . A. Mitchell, Angew.
Chem. 9S (1983) 706; Angew. Chem. I n f . Ed. Engl. 22 (1983) 674.
[4] J. R Blackborrow, D. Young: Metal-Vapor Sv,ithesis in Orgunometalk
Chemisrry, Spinger, Heidelberg 1979
[S] K. J. Klabunde. Chemi.srry of Free Atoms und Particles, Academic Press,
New York 1980.
[6] Handbook ofChemisfryandPhysics, 59th Edit. CRC, Cleveland, O H 1986.
[7] a) A. J. L. Hanlan, G. A. Ozin, Inorg. Chem. 16 (1977) 2848; 2857; b) M .
Moskovits, J. E. Hulse, J Chem. Soc. Furaday Trans. 1977. 471; c) M.
Angew. Chem. Int Ed. Engl. 29 (199#) 126- 137
Moskovits, J. E. Hulse, J. Chem. Phys. 66 (1977) 3988; d) K. J. Klabunde,
B. B. Anderson, M. Bader, L. J. Radonovich,J Am. Chem. Soc. lOO(1978)
1313.
[8] R. W. Zoellner, K. J. Klabunde, Chem. Rev. 84 (1984) 545, and references
cited therein.
[9] D. L. Williams-Smith, L. R. Wolf, P. S . Skell, J. Am. Chem. Soc. 94(1972)
4042.
[lo] R . Middleton, J. R. Hull, S . R. Simpson, C. H. Tomlinson, P. L. Timms, J
Chem. Soc. Dalton Trans. 1973, 120.
[ l l ] S D. Ittel, C. A . Tolman, J. Organomet. Chem. 172 (1979) C47;
Organometull~cs1 (1982) 1432.
[12] L J. Radonovich. M. W. Eyring, T. J. Groshens, K. J. Klabunde, J Am.
Chrm Soc 104 (1982) 2816.
[13] H. F Efner. D. E. Tevault. W. B. Fox, R. R. Smardzewksi, J. Organomet.
Chem. 146 (1978) 45.
[14] L. K. Beard, Jr.. M. P. Silvon, P. S . Skell, J. Organomet. Chem. 209(1981)
245.
[lS] G A. Ozin. C. G. Francis, H. X. Huber, M. Andrews, L. Nazar, J. Am.
Chem. Soc. 103 (1981) 2453.
[16] S . F. Parker. C. H. F. Peden, J. Oganomer. Chem. 272 (1984) 411.
[I71 P. D. Morand, C. G. Francs, Organometallics. 4 (1985) 1653.
[18] D. W. Ball, Z. H. Kafafi, R. H. Hauge, J L. Margrave, J Am. Chem. Soc.
/OH (1986) 6621.
[19] E. 0. Fischer, F. Rohrscheid, 2. Naturforsch. B 17 (1962) 483.
: . Grevels, D. Hess, J. Am. Chem. Soc. 106 (1984)
[20] H Fleckner, FW
2027.
[21] K. Jonas. E. Deffense, D. Habermann, Angew. Chem. 95 (1983) 729;
Angrn. Chem. I n t . Ed. Engl. 22 (1983) 716; Angew. Chem. Suppl. 1983,
1005.
[22] C. Elschenbroich, R. Mockel, Angew. Chem. 89 (1977) 908; Angew. Chem.
Int. Ed. Engl. 16 (1977) 870; E. P. Kiindig, P. L.. Timms,. J. Chem. SOC.
Chem. Commun. 1977,912; J. Chem. Soc. Dalton Trans. 1980, 991.
[23] E. L. Muetterties. J. R. Bleake, E. J. Wucherer, T. A. Albright, Chem. Rev.
82 (1982) 499.
[24] J. W. Hull, W. L. Gladfelter, Organometallics 3 (1984) 605.
[25] J. 0 . Albright, L. D. Brown, S . Datta, J. K. Kouba, S . S . Wreford, B. M.
Foxman, J. Am. Chem. Soc. 99 (1977) 5518; J. 0. Albright, S . Datta, B.
Dezube, J. K. Kouba, D. S . Marynick, S . S . Wreford, B. M. Foxman ibid.
101 (1979) 611.
[26] R J. Harper, U S Pat. 3 073 855 (January 15, 1963), Ethyl Corporation.
[27] T. A. Manuel. Inorg. Chem. 3 (1964) 1794.
[28] U Zenneck, W. Frank, Angew. Chem. 98 (1986) 806; ,4ngew. Chem. Int.
Ed. Engl. 25 (1986) 831.
[29] H. Schaufele. D. Hu, H. Pritzkow, U. Zenneck, Organometallics 8 (1989)
396.
[30] C. Tolxdorff, H. Pritzkow, U. Zenneck, unpublished.
[31] A. Funhoff, H. Schdufele, U. Zenneck, .
I
Organomet. Chem. 345 (1988)
331
[32] W Siebert. Angew. Chem. 97 (1985) 924; Angew. Chem. Int. Ed. Engl. 24
(1985) 943.
[33] U. Zenneck, L. Suber, H. Pritzkow, W. Siebert, Chem. Ber. 119 (1986)971.
[34] JLK. Uhm, H. Romich, H. Wadepohl, W. Siebert, 2. Naturforsch. B 42
(1988) 306.
[35] J:K. Uhm, H. Pritzkow, U. Zenneck, W. Siebert, unpublished.
[36] D. Buchner, Dissertation, Universitat Heidelberg 1986; H. Fuess, P.
Schweiss (Frankfurt am Main), H. Pritzkow (Heidelberg), unpublished.
[37] J. Edwin, M. C. Bohm, N. Chester, D. M. Hofman, R. Hoffmann. H.
Pritzkow. W. Siebert, K. Stumpf, H. Wadepohl, Organometallics 2 (1983)
1666.
[38] J. Zwecker, Dissertation, Universitat Heidelberg 1986.
[39] D. Buchner, T. Kuhlmann, G. Brodt, W. Siebert, unpublished.
[40] A. Funhoff, U. Zenneck, W. Siebert, unpublished.
[41] V. Schifer, H. Pritzkow, W. Siebert, Chem. Ber. 122 (1989) 467.
[42] G. Brodt. Dissertation, Universitat Heidelberg 1988.
[43] M. L. H. Green, L.-L. Wong, J Chem. Soc. Chem. Commun. 1984, 1442;
J. Chem. Soc. Dalton Trans. 1987,411; M. L. H. Green, D. O’Hare, L:L.
Wong, ibid. 1987, 2031.
[44] H Schaufele, Dissertation. Universitat Heidelberg 1989.
I
[45] B. Wdlther, Coord. Chem. Rev.60 (1984) 67; H. Neukomm, H. Werner, .
Organornet. Chem. 108 (1976) C26; H. Werner, R. Feser, Z . Anorg. Allg.
Chem. 458 (1979) 301; W. KIHui, Z . Nalurforsch. B 34 (1979) 1403; Q.-B.
Bdo. T. B. Brill, Organometatlics 6 (1987) 2588.
[46] D. H. M. W. Thewissen.1 Organomet. Chem. 192(1980) 115; H. P. M. M.
Ambrosius, J. H. Noordik, G. J. A. Ariaans, J. Chem. Soc. Chem. Commun. 1980, 832.
[47] S . D. Ittel, P. J. Krusic, P. Meakin, J. Am. Chem. Soc. 100 (1978) 3264.
1481 D. Hu, H. Pritzkow, H. Schaufele, C. Tolxdorff, U. Zenneck in H. Werner,
G Erker (Eds.): Organometallics in Organic Synthesis, Vol. 2, Springer,
Heidelberg 1989, p. 45.
[49] T. V. Harris. J. W. Rathke, E. L. Muetteries, J. Am. Chem. Soc. 100 (1978)
6966.
[SO] J L. Tempelton, B. C. Ward, J. Am. Chem. Soc. 102 (1980) 3288.
Angen. Chem. In1 Ed. Engl. 29 (1990) 126-137
1511 H. Schaufele, H. Pritzkow, U. Zenneck, Angew. Chem. 100 (1988) 1577;
Angew. Chem. In[. Ed. Engl. 27 (1988) 1519.
[52] J. Roziere (Montpellier), unpublished EXAFS-results.
[53] L. Vasquez, Dissertation, Universitat Heidelberg 1988; J. Ensling, P. Gutlich (Mainz), unpublished.
[54] J. R. Blackborrow, unpublished results cited as Ref. [65 b] in [8].
[55] P. L. Timms, Angew. Chem. 87(1975) 295; Angew. Chem. Int. Ed. Engl. 14
(1975) 273.
[56] Y. Wakatsuki, 0. Nomura, K. Kitaura, K. Marokuma, H. Yamazaki, J
Am. Chem. Soc. 105 (1983) 1907; Y Wakatsuki, H. Yamazaki, L Orgunomet. Chem. 139 (1977) 169; M. B. Gomesde Lima, J. E. Guerchals, F. J. Le
Floch-Ptrennou, P. L. L’Haridon, F. Y. Pttillon, J. Taparmin, Organometallics 6 (1987) 1930; D. W. Macomber, A. V. Verma, R. D. Rogers, ibid. 7
(1988) 1241.
[57] G. Becker, G. Gresser, W. Uhl, 2. Naturfbrsch. B 36 (1981) 16.
[58] P. B. Hitchcock, M. J. Maah, J. F. Nixon, J. Chem. Soc. Chem Commun.
1986, 737; P. Binger, R. Milczarek, R. Mynott, M. Regitz, W. Rosch,
Angew. Chem. 98 (1986) 645; Angeu? Chem. Int. Ed. Engl. 25 (1986) 644.
[59] M. Regitz, P. Binger, Angew. Chem. 100 (1988) 1541; Angew. Chem. lnt.
Ed. Engl. 27 (1988) 1484, and references cited therein.
[60] J. F. Nixon, Chem. Rev. 88 (1988) 1327. and references cited therein.
[61] M. Driess. D. Hu, H. Pritzkow, H. Schdufele, U. Zenneck, M. Regitz, W.
Rosch, J. Organornet. Chem. 334 (1987) C 3 5 .
[62] 36 and 37 are also accessible via another route: R. Bartsch, P. B. Hitchcock, J. F. Nixon, J. Chem. Soc. Chem. Commun. 1987, 1146.
[63] 0. J. Scherer, T. Briick, Angew. Chem. 99 (1987) 59; Angew. Chem. Int. Ed.
Engl. 26 (1987) 59; 0. J. Scherer, T. Briick, G. Wolmershauser, Chem. Ber.
121 (1988) 935.
[64] W. Rosch, U. Hees, M. Regitz, Chem. Ber. 120 (1987) 1645.
[65] D. Hu, H. Schaufele, H. Pritzkow, U. Zenneck, Angew. Chem. 101 (1989)
929; Angew. Chem. I n l . Ed. Engl. 28 (1989) 900.
[66] D. Hu, H. Schaufele, H. Pritzkow, U. Zenneck, unpublished.
[67] H. Hoberg, K. Jenni. K. Angermund. C. Kriiger, Angew. Chem. 99 (1987)
141; Angew. Chem. I n t . Ed. Engl. 26 (1987) 153.
[68] See also: A. Marinetti, F. Mathey, J. Fischer, A. Mitschler. J. Am. Chem.
Soc. 104 (1982) 4484; J. Chem. SOC.
Chem. Commun. 1984, 182.
[69] T. Wettling, J. Schneider, 0. Wagner. C. G. Kreiter, M. Regitz, Angew.
Chem. 101 (1989) 1035; Angew. Chem. I n t . Ed. Engl. 28 (1989) 1013.
[70] E. L. Allred, B. R. Beck, J Am. Chem. Soc. 95 (1973) 2393.
[71] E. C. Smith, J. C. Barborak, J. Org. Chem. 41 (1976) 1433; A. P. Marchand, T:C. Chou, J. D. Eckstrand, D. van der Helm, ibid. 41 (1976) 1438;
P. E. Eaton, L. Cassar, R. A. Hudson, R. H. Deng, ibid. 41 (1976) 1445.
[72] M. Baudler, Angew. Chem. 99 (1987) 429; Angew. Chem. Inc Ed. Engl. 26
(1987) 419; H. G. von Schnering, W. Honk, Chem. Rev. 88 (1988) 243, and
references cited therein.
[73] T. G. Groshens, B. Henne, D. E. Bartok, K. J. Klabunde, Inorg. Chem. 20
(1981) 3629.
[74] K. J. Klabunde, D. Ralston, R. Zoellner, H. Hattori, Y. Tanaka, J. Catal.
55 (1978) 213.
[75] G. Vitulli, S . Bertozzi, M. Vignali, R. Lazzaroni, P. Salvatori. J Organomet. Chem. 326 (1987) C33.
[76] L. Vasquez, H. Pritzkow, U. Zenneck, Angew. Chem. 100 (1988) 705;
Angew. Chem. i n t . Ed. Engl. 27 (1988) 706.
[77] R. D. W. Kemmit, D. R. Russell in G. Wilkinson, F. G. A. Stone (Eds.):
Komprehensive Organometallic Chemistry Vol. 5, Pergamon, Oxford, 1982,
p. 195.
[78] A second route to this class of substances has ben reported recently: P.
Biagini, T. Funaioli, G. Fachinetti. F. Laschi, P. F. Zanazzi, J. Chem. Soc.
Chem. Commun. 1989,405.
[79] K. J. Klabunde, H. F. Efner, T. 0. Murdock, R. Ropple J. Am. Chem. Soc.
98 (1976) 1021.
[80] K. J. Klabunde, H. F. Efner, J. Fluorine Chem. 4 (1974) 114; J. H. Sinfelt,
J Catal. 29 (1973) 308; K. J. Klabunde, S . Davis, H. Hattori, Y. Tanaka,
ibid. 54 (1978) 254.
[Sl] R. M. Atkins, R. Mackenzie, P. L. Timms, T. W. Turney, J. Chem. Soc.
Chem. Commun. 1975, 764.
[82] K. Fischer, K. Jonas, G. Wilke, Angew. Chem. 85 (1973) 620; Angew.
Chem. I n t . Ed. Engl. 12 (1973) 595.
[83] W. Herter, Dissertation, Universitat Heidelberg 1984.
1841 J. Zwecker, H. Pritzkow, U. Zenneck, W. Siebert, Angew. Chem. 98 (1986)
1129; Angew. Chem. l n t . Ed. Engl. 25 (1986) 1099.
(851 T. Kuhimann, H. Pritzkow, U. Zenneck, W. Siebert, Angew. Chem. 96
(1984) 994; Angew. Chem. I n t . Ed. Engl. 23 (1984) 371.
[86] J. Zwecker, T. Kuhlmann, H. Pritzkow, W. Siebert, U Zenneck. Organometallics 7 (1988) 2316.
(871 H. P. Fritz, K. E. Schwarzhans, J Organornet. Chem. 1 (1964) 208; E. 0.
Fischer, R. Jira, Z. Naturforsch. 8 8 (1953) 217; P. Prins, J. D.- van Voorst,
C. J. Schinkel, Chem. Phys. Lett. 1 (1967) 54.
[88] J. L. Robbins, N. Edelstein, B. Spencer, J. C. Smart, J. Am. Chem. Soc. 104
(1982) 1882.
[89] M. V. Rajasekharan, R. Bucher. E. Deiss, L. Zoller, A. K. Salzer, E. Moser, J. Weber, J. H. Ammeter, J Am. Chem. Soc. 105 (1983) 7516.
137
Документ
Категория
Без категории
Просмотров
1
Размер файла
1 161 Кб
Теги
nickell, reaction, vapor, intermediate, iron, cobalt, reactive, areneв, highly, cocondensation
1/--страниц
Пожаловаться на содержимое документа